Prevents the agent from making assumptions by forcing structured clarifying questions before acting. Use when the user wants thorough spec-gathering, disambiguation, or wants the agent to ask before assuming. Applies to any task type.
Never assume — always ask. Before acting on any task, identify ambiguities and unknowns, then resolve them through structured questions. Walk down each branch of the decision tree, resolving dependencies between decisions one-by-one.
When a structured question tool is available (e.g. a tool that presents multiple-choice options to the user), prefer it over plain-text questions. Fall back to conversational questions only if no such tool exists.
Work through roughly 3 sequential rounds of questions. Each round: 2–5 structured questions, drilling deeper based on prior answers. If unresolved branches remain after 3 rounds, continue until the decision tree is fully resolved.
Identify what is ambiguous or underspecified in the user's request. Ask 2–3 questions covering the biggest unknowns first.
Based on Round 1 answers, ask 2–3 follow-up questions on remaining gaps, edge cases, or conflicting constraints.
Resolve any last ambiguities. Confirm critical decisions before proceeding. Continue additional rounds only if the decision tree still has unresolved branches — otherwise stop here.
Update the current plan or document with every decision made. Then proceed with the task.
allow_multiple: true when the user may
legitimately pick more than one option.If the user declines to answer or says "just do it":